The rise of layer 2 networks was one of the key themes in the crypto space last year, but as blockchain interoperability and scalability come into even bigger focus, layer 3 networks might start to see some time in the spotlight.

Layer 2 networks are built on top of the underlying blockchain, offering cheaper and faster transactions by combining a bundle of off-chain transactions into a single layer 1 transaction, benefiting from its security. 

On the other hand, layer 3 networks submit transaction bundles to the layer 2 network, offering even lower gas fees given the significantly lower cost of storing transactions on a layer 2. 

To some, this makes layer 3 networks the ultimate scaling solution, particularly for hosting decentralized applications (dApps) and executing custom functions. However, not everyone is buying into that narrative.

“I’ll say the quiet part out loud: [layer 3s] exist only to take value away from Ethereum and onto the [layer 2s] on which the [layer 3s] are built,” said Polygon Labs CEO Marc Boiron on X. 

He further argued that in an extreme case where all layer 3 networks settled on to one layer 2 then Ethereum would capture “basically no value” and its security would be at risk. 

Polygon Labs is the entity behind prominent layer 2 networks, including the Polygon proof-of-stake blockchain and the Polygon zero-knowledge Ethereum Virtual Machine (zkEVM) rollup. 

“If Ethereum earns no fees and has no prospect of earning fees other than a tiny amount of fees from this one [layer 2], then the value of ether will drop and, when it is clear that it’ll continue to drop because there is no economic future for it, validators will no longer be willing to hold eth and, thus, no longer be willing to secure the network,” Boiron said.

Boiron’s comments come after the launch of Degen Chain, a layer 3 network built on the Ethereum layer 2 Base, designed specifically for the Farcaster community and the DEGEN token. So far, most of the chain’s ecosystem activity has been centered around memecoin trading.